CN103605956A - Emergency intelligence identification system based on video abstraction - Google Patents
Emergency intelligence identification system based on video abstraction Download PDFInfo
- Publication number
- CN103605956A CN103605956A CN201310536952.0A CN201310536952A CN103605956A CN 103605956 A CN103605956 A CN 103605956A CN 201310536952 A CN201310536952 A CN 201310536952A CN 103605956 A CN103605956 A CN 103605956A
- Authority
- CN
- China
- Prior art keywords
- video
- subsystem
- frequency abstract
- video frequency
- system based
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses an emergency intelligence identification system based on a video abstraction. The system comprises a computer and a plurality of cameras. A video abstraction subsystem, a firework detection subsystem, a pornography detection subsystem, a sensitive information detection subsystem and a banner identification subsystem are arranged in the computer. The plurality of cameras are connected with the computer. Through cooperatively using the video abstraction subsystem, the firework detection subsystem, the pornography detection subsystem, the sensitive information detection subsystem and the banner identification subsystem, by using a video abstraction technology and a video event analysis technology, abstraction events of the whole video can be rapidly extracted without knowing a concrete characteristic of the video event and the video event analysis technology is used to analyze the abstraction events so as to determine what types of events they belong to. Therefore, automatic, rapid and high-efficient effects of video monitoring are achieved.
Description
Technical field
The present invention relates to field of video monitoring technology, refer in particular to a kind of accident intelligent identifying system based on video frequency abstract.
Background technology
In recent years, along with the rapid progress of society and the continuous enhancing of national power, grow with each passing day to the demand of safety precaution and field notes warning system in the fields such as bank, electric power, traffic, safety check and military installations, require more and more highlyer, video monitoring has obtained very widely application in productive life each side.Although supervisory system is own through being present in widely the public places such as bank, market, station and traffic intersection, but actual monitor task still needs more manually completing, and existing video monitoring system is recorded video image conventionally, the information providing is the video image that process is not explained, can only, as evidence obtaining afterwards, not give full play to real-time and the initiative of monitoring.For can real-time analysis, tracking, differentiation monitored object, and when anomalous event occurs, point out, report ，Wei government department, the timely decision-making in security fields, correct action to provide support, " intellectuality " of video monitoring just seems particularly important.
In present field of video monitoring, often a monitor video just has tens hours, and monitoring camera is also innumerable, in general be difficult to so many human resources and monitor or check so huge video resource, therefore by video monitoring robotization, the part interested of intelligent extraction video and the classification of Intelligent Recognition Video Events can greatly reduce taking of human resources.This plays a part very large to improving monitoring efficiency and security protection etc.
The classification of current intelligent video process software comparison main flow has:
(1) object identification
Can distinguish the classification of mobile object, be car, or motorcycle or people or aircraft etc., and this is the basis of other identifications.
(2) identification of crossing the border
Artificial picture one diatom or curve on video pictures, can identify the behavior that object passes through this boundary.Such as the visual field is on a road, draw a bundle of lines road and be divided into two ends, it is from left to right legal supposing to have defined, and is illegal from right to left, once Vehicle Driving Cycle has been crossed over this boundary line, equipment judges that whether it is illegal, non-rule produces reports to the police.
(3) track following
After identification place mobile object, can after mobile element, draw its motion through the track in place.As public places such as square, stations, the stream of people shuttles back and forth, and equipment can show and record everyone track of walking about, if a people hovers and goes around for a long time in the visual field, surpasses certain hour, and suspicious actions personage is found in the prompting of equipment automatic alarm.
(4) leave over or lose object identification
Equipment can identify object in the scene of the visual field and have more one or lacked one, be applicable to the places such as warehouse, station, exhibition room, safety check, if nobody picks up to have knapsack to be lost in somewhere for a long time, surpass the time of setting, system will produce warning, or picture these places of exhibition room, if exhibit lacks one, equipment also can be found and be reported to the police.
(5) car plate identification
If crossing, video scene Shi Ge road or gateway, community; Need only license plate area and occurred in video, equipment can automatically identify the number-plate number; And point out user in the mode of word.Can be for violation vehicle auditing, such as certain licence plate vehicle is escaped and do not known place to go after accident, if Ge Yao road junction, city has intelligence to know video server, system only has by the disposable number-plate number that one or several need to be checked of network and is set in each smart machine in system, once the vehicle of this licence plate occurred in the visual field, just alarm immediately.Save many police strength resources.
(6) vehicle speed measurement
Such as there being the speed prompt district of 200M on expressway, remind constantly driver not drive over the speed limit, yet drive over the speed limit, still occur repeatedly; An intelligent video server is installed on expressway if just think, as long as we draw two diatoms in the visual field, if know that the actual range of this two diatom is 100M, is input in equipment, equipment just can calculate the speed that each entered vehicle automatically, and hypervelocity time is reported to the police immediately.
(7) traffic statistics
Smart machine can identify passing pedestrian and vehicle, can count passing people or the quantity of car simultaneously; Just think on the doorway of a crossroad or Yi Ge conference and exhibition center, so passing vehicle flowrate of smart machine ，Jiu Neng Statistics Division or flow of the people are installed.For bus dispatching provides more information more timely.
(8) retrograde alarm
Such as one-way road, or outlet or the entrance on station, airport, the wagon flow stream of people is unidirectional, once there be people to drive in the wrong direction, system can automatically identify, and produces and report to the police.
(9) scribble behavior identification
Originally clean metope, is disorderly pasted small advertisement, phone or coating and drawing at random etc. by people, hinders the appearance of the city.There is intelligent monitor system equipment, just can find in time such scribble behavior, and alarm in time.
(10) the abnormal behaviour video such as fight
The public situation such as park, square, station, the stream of people is numerous, and the behavior of the harm personal safety of any burst all may cause heavy losses and negative effect.Intelligent identifying system, in time the abnormal behaviour of finder or vehicle, runs suddenly, falls down, and the behavior such as pursues and attacks, and supvr can be found and remind to system all in time.
These Video Events that can only identification software will detect or object are all with determinacy above, there are the features such as certain color, shape, present subject matter is, if you and when not knowing the specific features of jobbie, will go a monitor video to check from the beginning to the end once in artificial mode, so very labor intensive resource and time resource.
Summary of the invention
In view of this, the present invention is directed to the disappearance of prior art existence, its fundamental purpose is to provide a kind of accident intelligent identifying system based on video frequency abstract, and it can effectively solve existing video monitoring labor intensive resource and the problem of time resource.
For achieving the above object, the present invention adopts following technical scheme:
A kind of accident intelligent identifying system based on video frequency abstract, include computing machine and a plurality of camera, in this computing machine, be provided with video frequency abstract subsystem, pyrotechnics detection subsystem, pornographic detection subsystem, sensitive information detection subsystem and banner recognition subsystem, the plurality of camera all connects computing machine.
Preferably, described video frequency abstract subsystem can simultaneously treated video frequency abstract event be at least 150, and all summary event video lengths that generate are in 2 minutes.
Preferably, described video frequency abstract subsystem is in video frequency abstract leaching process, and extraction time is less than 1/5th of whole video length.
Preferably, described pyrotechnics detection subsystem gave the alarm when encountering doubtful smog sign and doubtful flame sign at real-time monitor video in 1 second, and returned to the time that occurs doubtful fire alarm appearance in monitoring in real time.
Preferably, described pornographic detection subsystem is distinguished swimsuit, the male sex skin picture and baby's nude of exposed, the non-pornographic of large area above the waist.
Preferably, described banner recognition subsystem can give the alarm when encountering doubtful banner event at real-time monitor video in 1 second, and returned to the time that occurs doubtful banner appearance in monitoring in real time.
Preferably, described a plurality of camera is high definition Explosionproof camera.
The present invention compared with prior art has obvious advantage and beneficial effect, particularly, and as shown from the above technical solution:
By cooperation, utilize video frequency abstract subsystem, pyrotechnics detection subsystem, pornographic detection subsystem, sensitive information detection subsystem and banner recognition subsystem, video summarization technique and Video Events analytical technology do not need to know the specific features of Video Events, the summary event by whole video that can be very fast extracts and adopts Video Events analytical technology to analyze it, judge that they belong to respectively the event of what type, reach video monitoring robotization, rapid, the effect of high efficiency, the present invention utilizes computer vision technique to process vision signal, analyze and understand, in the situation that not needing human intervention, by sequence image automatic analysis is positioned the variation in monitoring scene, identification and tracking, and analyze on this basis and judge the behavior of target, can when abnormal conditions occur, give the alarm in time or useful information is provided, effectively assist Security Officer to process crisis, and reduce to greatest extent wrong report and fail to report phenomenon.
For more clearly setting forth architectural feature of the present invention and effect, below in conjunction with accompanying drawing and specific embodiment, the present invention is described in detail.
Accompanying drawing explanation
Fig. 1 is the one-piece construction schematic diagram of the present invention's preferred embodiment;
Fig. 2 is the main interface of video frequency abstract subsystem in the present invention's preferred embodiment;
Fig. 3 is the process flow diagram that in the present invention's preferred embodiment, video frequency abstract subsystem extracts;
Fig. 4 is pyrotechnics overhaul flow chart in the present invention's preferred embodiment;
Fig. 5 is the process flow diagram of porny identification in the present invention's preferred embodiment;
Fig. 6 is the processing flow chart of video information in the present invention's preferred embodiment;
Fig. 7 is banner overhaul flow chart in the present invention's preferred embodiment.
Accompanying drawing identifier declaration:
10, computing machine 11, video frequency abstract subsystem
12, pyrotechnics detection subsystem 13, pornographic detection subsystem
14, sensitive information detection subsystem 15, banner recognition subsystem
20, camera
Embodiment
Please refer to shown in Fig. 1 to Fig. 7, the concrete structure that it has demonstrated the present invention's preferred embodiment, includes computing machine 10 and a plurality of camera 20.
Wherein, in this computing machine 10, be provided with video frequency abstract subsystem 11, pyrotechnics detection subsystem 12, pornographic detection subsystem 13, sensitive information detection subsystem 14 and banner recognition subsystem 15, the plurality of camera 20 all connects computing machine 10, in the present embodiment, the plurality of camera 20 is high definition Explosionproof camera.
This video frequency abstract subsystem 11 can simultaneously treated video frequency abstract event be at least 150, and all summary event video lengths that generate are in 2 minutes.This video frequency abstract subsystem 11 is in video frequency abstract leaching process, and extraction time is less than 1/5th of whole video length.This pyrotechnics detection subsystem 12 gave the alarm when encountering doubtful smog sign and doubtful flame sign at real-time monitor video in 1 second, and returned to the time that occurs doubtful fire alarm appearance in monitoring in real time.This pornographic detection subsystem 13 is distinguished swimsuits, the male sex skin picture and baby's nude of exposed, the non-pornographic of large area above the waist.This banner recognition subsystem 15 can give the alarm when encountering doubtful banner event at real-time monitor video in 1 second, and returned to the time that occurs doubtful banner appearance in monitoring in real time.
Particularly, as shown in Figure 2, be the main interface of video frequency abstract subsystem 11.
1. " selection video file " button: select a local video file.
2. " select File folder " button: select a local folders that comprises a plurality of videos.
3. show selected file path.
4. " analysis video " button: click this button and carry out video analysis.
5. the progress bar of display analysis video.
6. display analysis result.
7. " event is chosen in broadcasting " button: play the event of choosing from list.
8. the button of " playing all events ": play all events.
9. the list box of display video summary event.
Video frequency abstract subsystem 11 can the abstract extraction using the motion event in video as video out also can be checked summary event for user.Intelligent video summary software is mainly divided into 2 large functions: video frequency abstract extracts the broadcasting of (video analysis) and video frequency abstract, as shown in Figure 3, is the process flow diagram of video frequency abstract subsystem 11 extractions.
As shown in Figure 4, be pyrotechnics overhaul flow chart, to the identification of pyrotechnics, be mainly that fundamental characteristics based on smog and flame image is realized.Fire is a kind of out of control burning, and the smog of initial fire disaster and flame are to grow out of nothing, and is time-dependent, and flame profile, area, radiation intensity in the same time all do not changing, and smog is also that to spread gradually continuous change large.Catch these features of fire disaster flame and smog to provide good feature foundation for pyrotechnics identification.
The essential characteristic of smog image is mainly manifested in following several aspect:
What 1. smog occurred has or not.While detecting in the early stage, for smog, can from video image, detect smogless appearance judges.
2. the dynamic of smog.The dynamic of smog is to diffuse to form due to random.It comprises scrambling and the diffusivity of shape.We combine detection smog scrambling and diffusivity.The smog color region extracting is detected, judge whether to have scrambling and diffusivity.Like this can be by false smoke mist (there is smog color, but be not smog) region filtering.The detection of behavioral characteristics will greatly improve accurate alarm rate.
3. the color characteristic of smog.The color of smog all changes within the specific limits, can judge whether certain region is smog color according to smog color gamut in RGB color space.
Meanwhile, the fundamental characteristics main manifestations of flame image is following several aspect:
1. the variation of area.After fire occurs, its flame is constantly developing, and along with the increase of the intensity of a fire, flame also constantly strengthens.Thereby from the picture characteristics of flame, flame area is the trend that continuous extendability increases;
2. the variation at edge.The edge variation of fire disaster flame and other high temp objects and light and the edge variation stabilizing the flame also have certain difference.Can utilize the variation of flame fringe to carry out fire differentiation.Flame fringe shake is the characteristic of fire disaster flame, and other high temp objects, light and to stabilize the flame edge more stable;
3. the color characteristic of flame.Flame color is all to change in certain scope, can judge whether certain region is flame color according to RGB color space Flame color gamut.
As shown in Figure 5, process flow diagram for porny identification, for porny, there is following several feature: (1) color characteristic, in porny the inside, the baring skin area of human body can be very large, by extracting human body complexion shared ratio in whole image, the features such as shape can be used as the foundation of identification porny; (2) shape of sensitive part, textural characteristics, judge that whether a secondary picture is the still whether exposed human body sensitive part of porny most critical, on the basis of detecting at area of skin color, profile detection and Local textural feature extraction are carried out in this region, are second foundations of identification porny.
The identification of porny is adopted to the method for svm sorter classification, first abundant sample is carried out to color, shape, Texture eigenvalue and extract, then the feature of these samples of repetition training, draws a rational svm sorter.With the complete sorter of training, other pictures are classified again, porny is separated with normal picture.
As shown in Figure 6, sensitive information detection subsystem 14 adopts following technology path:
(1) set up one-piece construction, i.e. the classification of information storage, the typing of information, information model storage, sensitive information associated video information etc.Adopt database (MySQL/SQL server) to carry out persistence preservation to data.
(2) owing to carrying the logo of sensitive information in detection, need to template logo, process in advance, store and set up index, therefore no matter in the Video processing stage, still as target, search for video, all need to carry out certain processing as extracted color histogram, calculate its hu square, obtain some characteristics of image such as profile of template.
(3) UI design sketch
Can manually add new sensitive information.
(4) first, multitude of video or picture are taken in the middle of the system that is injected into by video camera, or direct and extraneous video source is set up contact.And in database, preserve relevant information.Then system is analyzed the fine granularity decomposition software of writing by system to video, by the image processing techniques storehouse of OpenCV, the eigenwert of the content of traffic video is extracted.Then create multimedia copy, and according to the eigenwert of extracting, multimedia copy is cut apart and compressed.This operation is convenient to improve the speed of search in the future.Meanwhile, machine files former multimedia to classify to unify to preserve according to certain rule with the multimedia of processing and manages and set up index.When user submits the request of some search to, the information of preserving by database to data through row access control.
As shown in Figure 7, be banner overhaul flow chart, banner have following several feature: (1) background is that pure color is (pure red, pure white, gilvous etc.), above banner, be then some posters or field, we can extract the profile of whole banner according to its background colour; (2) overall profile of banner is rectangle or class rectangular profile, and length breadth ratio is very greatly different, can after the rectangular profile of having extracted banner, compare by its length and width again, if be greater than 5:1, can be judged to be banner; (3) to only have background colour be pure color to banner, but whole banner is not pure color, because there is word, so just can filter out some pseudo-banners (whole colors of pseudo-banner are all the same).
Design focal point of the present invention is: by cooperation, utilize video frequency abstract subsystem, pyrotechnics detection subsystem, pornographic detection subsystem, sensitive information detection subsystem and banner recognition subsystem, video summarization technique and Video Events analytical technology do not need to know the specific features of Video Events, the summary event by whole video that can be very fast extracts and adopts Video Events analytical technology to analyze it, judge that they belong to respectively the event of what type, reach video monitoring robotization, rapid, the effect of high efficiency, the present invention utilizes computer vision technique to process vision signal, analyze and understand, in the situation that not needing human intervention, by sequence image automatic analysis is positioned the variation in monitoring scene, identification and tracking, and analyze on this basis and judge the behavior of target, can when abnormal conditions occur, give the alarm in time or useful information is provided, effectively assist Security Officer to process crisis, and reduce to greatest extent wrong report and fail to report phenomenon.
On know-why of the present invention has been described in conjunction with specific embodiments.These are described is in order to explain principle of the present invention, and can not be interpreted as by any way limiting the scope of the invention.Explanation based on herein, those skilled in the art does not need to pay performing creative labour can associate other embodiment of the present invention, within these modes all will fall into protection scope of the present invention.
Claims (7)
1. the accident intelligent identifying system based on video frequency abstract, it is characterized in that: include computing machine and a plurality of camera, in this computing machine, be provided with video frequency abstract subsystem, pyrotechnics detection subsystem, pornographic detection subsystem, sensitive information detection subsystem and banner recognition subsystem, the plurality of camera all connects computing machine.
2. the accident intelligent identifying system based on video frequency abstract as claimed in claim 1, is characterized in that: described video frequency abstract subsystem can simultaneously treated video frequency abstract event be at least 150, and all summary event video lengths that generate are in 2 minutes.
3. the accident intelligent identifying system based on video frequency abstract as claimed in claim 1, is characterized in that: described video frequency abstract subsystem is in video frequency abstract leaching process, and extraction time is less than 1/5th of whole video length.
4. the accident intelligent identifying system based on video frequency abstract as claimed in claim 1, it is characterized in that: described pyrotechnics detection subsystem gave the alarm when encountering doubtful smog sign and doubtful flame sign at real-time monitor video in 1 second, and returned to the time that occurs doubtful fire alarm appearance in monitoring in real time.
5. the accident intelligent identifying system based on video frequency abstract as claimed in claim 1, is characterized in that: described pornographic detection subsystem is distinguished swimsuit, the male sex skin picture and baby's nude of exposed, the non-pornographic of large area above the waist.
6. the accident intelligent identifying system based on video frequency abstract as claimed in claim 1, it is characterized in that: described banner recognition subsystem can give the alarm when encountering doubtful banner event at real-time monitor video in 1 second, and return to the time that occurs doubtful banner appearance in monitoring in real time.
7. the accident intelligent identifying system based on video frequency abstract as claimed in claim 1, is characterized in that: described a plurality of cameras are high definition Explosionproof camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310536952.0A CN103605956A (en) | 2013-11-04 | 2013-11-04 | Emergency intelligence identification system based on video abstraction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310536952.0A CN103605956A (en) | 2013-11-04 | 2013-11-04 | Emergency intelligence identification system based on video abstraction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103605956A true CN103605956A (en) | 2014-02-26 |
Family
ID=50124176
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310536952.0A Pending CN103605956A (en) | 2013-11-04 | 2013-11-04 | Emergency intelligence identification system based on video abstraction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103605956A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104506804A (en) * | 2014-12-22 | 2015-04-08 | 安徽三联交通应用技术股份有限公司 | Device and method for monitoring abnormal behavior of motor vehicle on expressway |
| CN106911550A (en) * | 2015-12-22 | 2017-06-30 | 腾讯科技（深圳）有限公司 | Information-pushing method, information push-delivery apparatus and system |
| CN110210356A (en) * | 2019-05-24 | 2019-09-06 | 厦门美柚信息科技有限公司 | A kind of picture discrimination method, apparatus and system |
| WO2020134677A1 (en) * | 2018-12-26 | 2020-07-02 | 中兴通讯股份有限公司 | Unlawful advertisement processing method and apparatus, and computer-readable storage medium |
| CN112368750A (en) * | 2018-06-04 | 2021-02-12 | 豪倍公司 | Emergency notification system |
| CN113205015A (en) * | 2014-04-08 | 2021-08-03 | 乌迪森斯公司 | System and method for configuring a baby monitor camera |
-
2013
- 2013-11-04 CN CN201310536952.0A patent/CN103605956A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113205015A (en) * | 2014-04-08 | 2021-08-03 | 乌迪森斯公司 | System and method for configuring a baby monitor camera |
| CN104506804A (en) * | 2014-12-22 | 2015-04-08 | 安徽三联交通应用技术股份有限公司 | Device and method for monitoring abnormal behavior of motor vehicle on expressway |
| CN106911550A (en) * | 2015-12-22 | 2017-06-30 | 腾讯科技（深圳）有限公司 | Information-pushing method, information push-delivery apparatus and system |
| CN112368750A (en) * | 2018-06-04 | 2021-02-12 | 豪倍公司 | Emergency notification system |
| US11570872B2 (en) | 2018-06-04 | 2023-01-31 | Hubbell Lighting, Inc. | Emergency notification system |
| WO2020134677A1 (en) * | 2018-12-26 | 2020-07-02 | 中兴通讯股份有限公司 | Unlawful advertisement processing method and apparatus, and computer-readable storage medium |
| CN110210356A (en) * | 2019-05-24 | 2019-09-06 | 厦门美柚信息科技有限公司 | A kind of picture discrimination method, apparatus and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103605956A (en) | Emergency intelligence identification system based on video abstraction | |
| CN104700434B (en) | A kind of crowd movement track method for detecting abnormality for labyrinth scene | |
| CN109002744A (en) | Image-recognizing method, device and video monitoring equipment | |
| CN103310183B (en) | The method and apparatus of crowd massing detection | |
| CN100583128C (en) | Real time intelligent control method based on natural video frequency | |
| CN102819764B (en) | Method for counting pedestrian flow from multiple views under complex scene of traffic junction | |
| CN103116989B (en) | Based on the catching method of the fake-licensed car vehicle of smart city | |
| CN103347167A (en) | Surveillance video content description method based on fragments | |
| Lin et al. | Learning to detect anomaly events in crowd scenes from synthetic data | |
| CN102880692A (en) | Retrieval-oriented monitoring video semantic description and inspection modeling method | |
| CN103413330A (en) | Method for reliably generating video abstraction in complex scene | |
| CN103069434A (en) | Multi-mode video event indexing | |
| KR20060031832A (en) | A smart visual security system based on real-time behavior analysis and situation cognizance | |
| CN107346435A (en) | A kind of suspicion fake-licensed car catching method based on vehicle characteristics storehouse | |
| CN105426820A (en) | Multi-person abnormal behavior detection method based on security monitoring video data | |
| CN104811655A (en) | System and method for film concentration | |
| CN109872541A (en) | A kind of information of vehicles analysis method and device | |
| CN104637071A (en) | People tracking method based on intelligent video analysis | |
| CN107277470A (en) | A kind of network-linked management method and digitlization police service linkage management method | |
| Qiang et al. | Forest fire smoke detection under complex backgrounds using TRPCA and TSVB | |
| CN108965804A (en) | A kind of video structural technology for city security protection | |
| CN105657372A (en) | Method and system for realizing intelligent detection and early warning of on-duty guard posts by videos | |
| CN104469299A (en) | Network camera shooting device | |
| Hampapur et al. | Searching surveillance video | |
| CN114885119A (en) | Intelligent monitoring alarm system and method based on computer vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140226 |
|
| WD01 | Invention patent application deemed withdrawn after publication |